For the problems occurring in a least square method model, a fuzzy model, and a neural network model for flatness pattern recognition, a fuzzy neural network model for flatness pattern recognition with only three-inpu...For the problems occurring in a least square method model, a fuzzy model, and a neural network model for flatness pattern recognition, a fuzzy neural network model for flatness pattern recognition with only three-input and three output signals was proposed with Legendre orthodoxy polynomial as basic pattern, based on fuzzy logic expert experiential knowledge and genetic-BP hybrid optimization algorithm. The model not only had definite physical meanings in its inner nodes, but also had strong self-adaptability, anti interference ability, high recognition precision, and high velocity, thereby meeting the demand of high-precision flatness control for cold strip mill and providing a convenient, practical, and novel method for flatness pattern recognition.展开更多
To adapt to the new requirement of the developing flatness control theory and technology, cubic patterns were introduced on the basis of the traditional linear, quadratic and quartic flatness basic patterns. Linear, q...To adapt to the new requirement of the developing flatness control theory and technology, cubic patterns were introduced on the basis of the traditional linear, quadratic and quartic flatness basic patterns. Linear, quadratic, cubic and quartic Legendre orthogonal polynomials were adopted to express the flatness basic patterns. In order to over- come the defects live in the existent recognition methods based on fuzzy, neural network and support vector regres- sion （SVR） theory, a novel flatness pattern recognition method based on least squares support vector regression （LS-SVR） was proposed. On this basis, for the purpose of determining the hyper-parameters of LS-SVR effectively and enhan- cing the recognition accuracy and generalization performance of the model, particle swarm optimization algorithm with leave-one-out （LOO） error as fitness function was adopted. To overcome the disadvantage of high computational complexity of naive cross-validation algorithm, a novel fast cross-validation algorithm was introduced to calculate the LOO error of LDSVR. Results of experiments on flatness data calculated by theory and a 900HC cold-rolling mill practically measured flatness signals demonstrate that the proposed approach can distinguish the types and define the magnitudes of the flatness defects effectively with high accuracy, high speed and strong generalization ability.展开更多
In the traditional flatness pattern recognition neural network, the topologic configurations need to be rebuilt with a changing width of cold strip. Furthermore, the large learning assignment, slow convergence, and lo...In the traditional flatness pattern recognition neural network, the topologic configurations need to be rebuilt with a changing width of cold strip. Furthermore, the large learning assignment, slow convergence, and local minimum in the network are observed. Moreover, going by the structure of the traditional neural network, according to experience, the model is time-consuming and complex. Thus, a new approach of flatness pattern recognition is proposed based on the CMAC （cerebellar model articulation controllers） neural network. The difference in fuzzy distances between samples and the basic patterns is introduced as the input of the CMAC network. Simultaneously, the adequate learning rate is improved in the error correction algorithm of this neural network. The new approach with advantages, such as high learning speed, good generalization, and easy implementation, is efficient and intelligent. The simulation results show that the speed and accuracy of the flatness pattern recognition model are obviously im proved.展开更多
基金Item Sponsored by National Natural Science Foundation of China and Shanghai Baosteel Group Co(50675186)Provincial Natural Science Foundation of Hebei Province of China(E2006001038)
文摘For the problems occurring in a least square method model, a fuzzy model, and a neural network model for flatness pattern recognition, a fuzzy neural network model for flatness pattern recognition with only three-input and three output signals was proposed with Legendre orthodoxy polynomial as basic pattern, based on fuzzy logic expert experiential knowledge and genetic-BP hybrid optimization algorithm. The model not only had definite physical meanings in its inner nodes, but also had strong self-adaptability, anti interference ability, high recognition precision, and high velocity, thereby meeting the demand of high-precision flatness control for cold strip mill and providing a convenient, practical, and novel method for flatness pattern recognition.
基金Sponsored by National Natural Science Foundation of China (50675186)
文摘To adapt to the new requirement of the developing flatness control theory and technology, cubic patterns were introduced on the basis of the traditional linear, quadratic and quartic flatness basic patterns. Linear, quadratic, cubic and quartic Legendre orthogonal polynomials were adopted to express the flatness basic patterns. In order to over- come the defects live in the existent recognition methods based on fuzzy, neural network and support vector regres- sion （SVR） theory, a novel flatness pattern recognition method based on least squares support vector regression （LS-SVR） was proposed. On this basis, for the purpose of determining the hyper-parameters of LS-SVR effectively and enhan- cing the recognition accuracy and generalization performance of the model, particle swarm optimization algorithm with leave-one-out （LOO） error as fitness function was adopted. To overcome the disadvantage of high computational complexity of naive cross-validation algorithm, a novel fast cross-validation algorithm was introduced to calculate the LOO error of LDSVR. Results of experiments on flatness data calculated by theory and a 900HC cold-rolling mill practically measured flatness signals demonstrate that the proposed approach can distinguish the types and define the magnitudes of the flatness defects effectively with high accuracy, high speed and strong generalization ability.
基金National Natural Science Foundation of China (50675186)
文摘In the traditional flatness pattern recognition neural network, the topologic configurations need to be rebuilt with a changing width of cold strip. Furthermore, the large learning assignment, slow convergence, and local minimum in the network are observed. Moreover, going by the structure of the traditional neural network, according to experience, the model is time-consuming and complex. Thus, a new approach of flatness pattern recognition is proposed based on the CMAC （cerebellar model articulation controllers） neural network. The difference in fuzzy distances between samples and the basic patterns is introduced as the input of the CMAC network. Simultaneously, the adequate learning rate is improved in the error correction algorithm of this neural network. The new approach with advantages, such as high learning speed, good generalization, and easy implementation, is efficient and intelligent. The simulation results show that the speed and accuracy of the flatness pattern recognition model are obviously im proved.
